import numpy as np
import wfdb as wf
import sys
import matplotlib.pyplot as plt
import tensorflow as tf
import os, imageio
from tqdm import tqdm
import mitecg
ECG = mitecg.ReadMitEcg(
'/Users/chen/Desktop/ecg/www.physionet.org/physiobank/database/mitdb',
10000, [1, 2, 3, 4, 5], True)
batch_size = 20
HEARTBEATSAMPLES = 250
LABEL = 5
z_dim = 100
X = tf.placeholder(dtype=tf.float32,
shape=[None, HEARTBEATSAMPLES, 1],
name='X')
y_label = tf.placeholder(dtype=tf.float32,
shape=[None, HEARTBEATSAMPLES, LABEL],
name='y_label')
noise = tf.placeholder(dtype=tf.float32, shape=[None, z_dim], name='noise')
y_noise = tf.placeholder(dtype=tf.float32, shape=[None, LABEL], name='y_noise')
is_training = tf.placeholder(dtype=tf.bool, name='is_training')
def lrelu(x, leak=0.2):
return tf.maximum(x, leak * x)
import numpy as np
import wfdb as wf
import sys
import matplotlib.pyplot as plt
import tensorflow as tf
import os, imageio
from tqdm import tqdm
import mitecg
from scipy.fftpack import fft, ifft
import seaborn
SCALEDSAMPLES = 50
ECG = mitecg.ReadMitEcg(
'/Users/chen/Desktop/Research/ecg/www.physionet.org/physiobank/database/mitdb',
10000, [1, 2, 3, 4, 5],
False,
SCALEDSAMPLES=SCALEDSAMPLES)
sampleArray = ECG.oneEcgWithHeartBeatScaled()
x = np.linspace(0, 1, sampleArray.size)
yy = fft(sampleArray)
yreal = yy.real
yimag = yy.imag
yf = abs(yy)
yf1 = abs(yy) / len(x)
yf2 = yf1[range(int(len(x) / 2))]
import tensorflow as tf
import os, imageio
from tqdm import tqdm
import mitecg
import argparse
parser = argparse.ArgumentParser()
parser.add_argument("datapath", help="directory to the mit ecg database")
args = parser.parse_args()
batch_size = 250
HEARTBEATSAMPLES = 50
LABEL = 5
z_dim = 100
ECG = mitecg.ReadMitEcg(args.datapath,
10000, [1, 2, 3, 4, 5],
True,
SCALEDSAMPLES=HEARTBEATSAMPLES)
X = tf.placeholder(dtype=tf.float32,
shape=[None, HEARTBEATSAMPLES, 1],
name='X')
y_label = tf.placeholder(dtype=tf.float32,
shape=[None, HEARTBEATSAMPLES, LABEL],
name='y_label')
noise = tf.placeholder(dtype=tf.float32, shape=[None, z_dim], name='noise')
y_noise = tf.placeholder(dtype=tf.float32, shape=[None, LABEL], name='y_noise')
is_training = tf.placeholder(dtype=tf.bool, name='is_training')
def lrelu(x, leak=0.2):
return tf.maximum(x, leak * x)